Online Plots

From PREX Wiki
Revision as of 16:59, 23 July 2019 by Rom (Talk | contribs) (What are our main detectors, and what happened to the rest of the HRS?)

Jump to: navigation, search

In the process of editing and updating this page. Please contact Juliette Mammei with questions or suggestions.


See Also: Alarm Documentation

For prompt analysis web-plots see here


Measuring, Controlling and Removing Helicity Correlated Noise in our Main Detectors

We assume that our main detectors are simultaneously measuring the helicity correlated motion and fluctuations of the electron beam from CEBAF, as well as physically interesting Parity Violating Asymmetry (Apv) of highly longitudinally-polarized electrons on an isotopically pure Lead 208 (Calcium 48) target of appreciable, but intentionally small scattering length.

The way that we can disentangle the beam position, angle, current and energy fluctuations due to helicity correlated beam (i.e. generated in the polarized source non-uniformly between helicity states and then damped or amplified farther along the beamline) are myriad.

Live and Online Analysis Result Plots

PREX II and CREX have a copy of the QWeak analyzer, referred to as JAPAN (Just Another Parity ANalyzer) is able to attach to our CODA data acquisition system and decode and analyze the data on the fly. We can attach to the live updating ROOT file from the online JAPAN analyzer and show live plots in the live root file plotting utility Panguin.

Critical Systems

There are several elements of the experiment which must be functioning properly in order for the data to be useful. The Parity Alarms keep track of a bunch of these, but also we can track a number of them in the online plots.

List of Critical Systems and Descriptions

Beam Current Monitor

  • BCMs measure the current entering the hall. They should be reading less than 90 uA at all times. This is important to not destroy our target or generate excess ion chamber trips.
  • The BCM is the object of scrutiny for our Charge Feedback (Aq) process, which is actively measuring the BCM asymmetries and sending information to the injector PITA system to reduce the charge asymmetry.
    • The goal is to prevent the charge asymmetry from reaching > 0.5 ppm levels, where our systematic errors come into play and make our charge normalization corrections less reliable
    • In the online plots you should monitor the bcm_an_ds3 analysis plots after doing the prompt analysis to ensure that all of these components are playing nicely together
  • There are two BCMs in Hall A, the upstream and downstream, and we prefer the downstream.
    • The upstream BCM should have a double difference w.r.t. downstream (difference of the two asymmetries) of ~15 to 20 ppm, and if this gets much larger or shows non-Gaussian behavior then the system is having problems and the RC and experts need to know about it immediately.
    • This double difference plot is also shown in the online plots

The BCM of interest at this time is the analog, downstream, x3 (bcm_an_ds3)

Hall A Arc, Energy Degree of Freedom-Sensitive BPM 12

This BPM is located in the BSY and measures beam position. It's also most sensitive to beam energy fluctuations. If the wires in the BPM saturate then its indicative of unacceptably high energy fluctuations. Therefore the readout must be kept below 50,000 at all times.

Missed Triggers in the DAQ

Main Detectors

What are our main detectors, and what happened to the rest of the HRS?

In each spectrometer we have two Quartz bars which intercept the beam. The Cerenkov signal integrated over the helicity is our main detector signal. We have two bars in each HRS for redundancy.

Also, we have so-called A_T detectors, which are also quartz bars. They are located at strategic locations in the detector frame so as to intercept electrons which have a maximum ``figure of merit for measuring a possible transverse asymmetry systematic.

The beamline elements are also important, the BCMs are used to normalize the quartz bars and the beam position monitors are used to regress (subtract) beam-related motion and to evaluate systematic errors associated with possible helicity correlations in the beam parameters like position, angle, and energy.

List of kinds of plots we care to make from main detectors

Raw Signal Voltage into the DAQ

Yield (signal in Volts/Microamps of Beam Current)

This measures our detector response as a function of beam current, with higher current generating stronger signal. Signals should be roughly the same for both left and right HRS's.


Small Angle Monitors (SAMs)

What are SAMs?

The SAMs consist of eight small pieces of scintillating quartz placed in the beamline roughly seven meters downstream of the target. Placed into the beamline, circularly at 45 degree intervals, these act as luminosity monitors, measuring both flux and scattering cross-section of the beam after hitting the lead target. PREX requires high luminosity to achieve the statistic necessary to observe the ppm-level asymmetries required.

Lists of kinds of plots

Detectors Yields (not-normalized to beam current)

SAM signal vs event number. Will read higher values at higher beam current. If signal is zero, check first whether or not SAM HV is off.

SAM Asymmetries

SAM asymmetries are plotted first by themselves, and then correlated with BCM asymmetry.

SAM beam profile/halo sensitivities around the circle/combining 8 Degrees of Freedom

SAM signals and asymmetries are plotted in a circle with the center plot indicating which SAM number corresponds to which SAM position as seen looking downstream along the beam line. Positional differences in the SAM readout may correspond to excess beam halo or give information about the beam profile. They may also act as a diagnostic for the detector expert crew if SAMs are behaving incorrectly.

Regressed SAMs as a noise floor limit indication (Advanced Plots)

Beam Position Monitors (BPMs)

What is a BPM and how does it work?

As the name suggests: a Beam Position Monitor monitors the beam position. A BPM consists of four wires arranged along the beam line

BPMs as multi-degree of freedom Beam-Position-Correlated noise regression monitors

Continuation of Measuring, Controlling and Removing Helicity Correlated Noise in our Main Detector Measurements

Using beam position monitors, to keep track of the momentum/energy fluctuations in the magnetically energy selecting Hall A arc, as well as the 4 degrees of X and Y position and angle onto the target information, allows us to measure helicity correlated beam fluctuations to fairly high precision and then calculate correlation slopes over various time scales and modes of beam motion (normal beam motion and intentionally large modulation induced beam motion) which are then subtracted, per event, from our main detectors to arrive at the remaining physics asymmetry.

List of plots of BPM related measurements

Calculated position yield

Beam Position on the target
Abstractly: Beam Angle on the Target
More Abstractly: Beam position as a measure of energy fluctuations in incident electrons

The same, but their multiplet/helicity state averaged differences

The same, but w.r.t. other detectors (SAMs, Mains, AT)

Beam Current Monitors (BCMs)

What is an Unser BCM and how does it work?

What is an analog/normal BCM and how does it work?

What is a digital BCM and how does it work?

What is a cavity BCM/BPM and how does it work?

List of BCM plots we care about

The Beam Current, Micro Amps

Beam charge Asymmetry

Beam charge Asymmetry, correlations with SAMs

Beam charge Asymmetry, correlations with already BCM normalized Main Detectors

Beam Modulation (BMW) Cycle/Response Plots

What is BMW and how does it work?

List of BMW plots

BMW Cycle Heartbeat

Beam Monitors vs. BMW Cycle Number/Phase

DAQ Timing and Synchronization Plots

Our Data Acquisition system (DAQ) is comprised of 5 VME crates and several NIM crates. There is one VME crate in the injector, counting house, and in each HRS hut. These crates are synchronized to all integrate the same time windows and read out the integrated signals at the same time and report them to the Trigger Supervisor (TS) VME crate and the CODA system for online analysis and storage of raw data to disk. We have several redundant systems in place to ensure that the various DAQ components are all measuring exactly the same data and to alert us when anything changes and by how much.

What is the DAQ Timing?

In general the DAQ timing is very precisely defined and comprised of a mixture of external trigger source from the helicity control board, gate generators, and internal clock based timing. The DAQ timing is documented fully and updated regularly in DAQ Layouts.

  • As mentioned above, it is important to ensure that the DAQ is measuring the same exact time information in all places, this is to guarantee that the helicity correlated signals used for determining our parity violating asymmetry and all of the corrections we must apply are regarding the same physical electrons across the integration window.
  • It is also important to ensure that our timing is safely defined to avoid such effects as Pockells Cell ringing, DAQ bandwidth limitations, and DAQ reinitialization trigger failure.
    • The Pockells Cell in the injector's polarized source is responsible for generating the longitudinally polarized and 120 (or 240) Hz flipping state electron beam. As there is no such thing as a perfect and instantaneous transition of macroscopic states, there is ringing of the Pockells Cell, where the flipped state changes to and oscillates around the stable desired state. Because this ringdown is affecting the beam polarization and can affect the normalization at a similar scale as our proposed asymmetry measurement, it is important to wait to begin our integration until the polarization state is stable (hence the name of the integration window "Tstable"). We use the complement of Tstable -> Tsettle to define when it is not safe to integrate and when it is best to begin the DAQ readout routines.
    • Because of the complex impedance in our DAQ electronics chain and integrating modules there is some finite bandwidth capacity, where a signal can decay over many microseconds, leading to an increase of the relative unsafe "Tsettle" time beyond just that related to the intrinsic instability of the beam due to Pockells Cell ringing. This is why Tsettle is longer than the ~40 us needed by the Pockells Cell alone.
    • An additional constraint on DAQ timing arises from the choice we have made to base our ADC and Scaler triggers on a programmable "Happex Timing Board" (haptb, or timebrd). This timing board was designed for the prior HAPPEX set of experiments and was responsible for defining the integration gate of the old (and still implemented as backup) HAPPEX ADCs. We have chosen to use the same logical integration gates as are used in the HAPPEX ADCs to also trigger and gate our QWeak ADCs (the main ADCs for all of our monitors) and scalers (for logical, less important, or V2F'd signals). The Happex Timing Boards have to be triggered by a signal at least 22.5 us before the time you desire the timebrd to begin its output gate routines (on a NIM/Lemo output signal called GMN2). Additionally the gate generated by the timebrd must not come within ~30 us of the following timebrd trigger. Hence we stop our integration gate several dozen us early before the end of Tstable.

What is the DAQ Synchronization System?

Kinds of Timing and Synchronization plots we care about

DAQ Integration Window Verification

Jitter in the end of a Helicity Multiplet

Calculated helicity flip rate

Missed Triggers

Synchronization and timing offset of Counting House w.r.t. HRSs and Injector

Phase Monitor

What is a phase monitor and why is it useful

Phase monitor by itself

Phase monitor correlations with other quantities as a canary in the coal mine

EPICs Archive Plots

Fast and Parity Feedback

Counting Mode DAQ Plots